210
Kupzok, Sturny-Luder and Surblyte
IICVoI.41
2/2010
Foundations and Limitations of an Economic Approach
211
chilling signal that such acquisitions will not be deemed favourable by
China. The controversial approach represents a quasi-judicial way to fend
off foreign acquirers. Conversely, the rejection could jeopardise Chinese
overseas expansion efforts and Chinese companies seeking outbound M&As
may encounter an adverse reaction.
2 the reform
Starting with the field of restrictive agreements (Art. 81 EC),
4 and,
3 the abuse of market dominance (Art. 82 EC)
covered merger control,
5 The gist of the reform was to shift the focus from a “form
finally, State aid.
based“ to an “effects-based“ approach and to make die analysis of the
allegedly anti-competitive conduct more appropriate.
The lack of instructive analytical standards and expianation of the reasoning
behind the prohibition may raise practical difficulties and render any future
review less predictable. More detailed analysis that would provide weil
needed guidance and transparency is highly expected in view of China‘s lack
of experience in handling sophisticated antitrust cases. In this vein, the new
anti-monopoly regime‘s prospects are not regarded as promising as there is
suspicion for its effective implementation and enforcement. Realistically, the
enforcement practices of MOFCOM are only in its infancy and understand
ably cannot compete with its more experienced counterparts in jurisdictions
such as the US and EU.
However, though the introduction of economic analysis into competition
iaw can heip better understand how markets function, the application of a
“more economic approach“ is not unproblematic. The involvement of eco
nomic assessment reduces legal certainty. lt requires decent economic knowl
edge from both competition law enforcers as weil as undertalcings. Difficul
ties also occur due to the use of different economic toois or models and
changing economic theories, which further raise unpredictability. All this
argues in favour of a cautious application of the “more economic approach“
in competition law.
The “Foundations and Limitations of an Economic Approach to Competition
Law“ were analysed at a conference held on 12 and 13 March 2009 at the
Max Planck Institute for Intellectual Property, Competition and Tax Law in
Munich. Co-hosted by Josef Drexl (Director at the Max Planck Institute,
Munich and Honorary Professor at the Ludwig Maximilian University of
Munich) and Wolfgang Kerber (Professor at Philipps University, Marburg),
the conference was devoted to the criticai assessment of the current and future
capabilities of a “more economic approach“. The issues were addressed
through the presentation of papers in six panels and following discussions.
Report
Agnieszka Kupzok, * Monique SturnyLuder* * and Gintar
Surblyt* * *
Foundations and Limitations of an Economic
Approach to Competition Law Conference of the
Max Planck Institute for Intellectual Property,
Competition and Tax Law, March 2009
—
Since the adoption of die European Commission Regulation on the applica
tion of Article 81(3) of the Treaty to categories of vertical agreements and
concerted practices,
1 die different areas of European competition law have
been constantly reformed under the so-cafled “more economic approach“.
LL.M IP (MIPLC); Ph.D. candidate at the International Max Planck Research School for
Competition and Innovation (IMPRS-CI); scholarship holder at the Max Planck Institute
for Intellectual Property, Competition and Tax Law, Munich, Germany.
**
Attorney-at-Law (Switzerland); LL.M. (LSE, London); Ph.D. candidate at the Institute
for European and International Economic Law, University of Berne, Switzerland.
LL.M. (Munich); Ph.D. candidate at the Max Planck Institute for Intellectual Property,
Competition and Tax Law, Munich, Germany.
1 Commission Regulation (EC) No. 2790/1999 of 22 December 1999, [1999] 0J L 336/
21.
*
Panel 1: The State of Affairs
Revisited
—
Modern Industrial Economics
Presentations
Michele Polo, Professor of Economics at the Bocconi University in Milan,
gave a presentation on the anticompetitive versus competitive explanations
2 Guidelines of the Commission on the application of Art. 81(3) of the Treaty, [20041 0J
C 101/97; Commission Regulation (EC) No. 772/2004 of 27 April 2004 on the applica
tion of Art. 81(3) of the Treaty to categories of technology transfer agreements, [20041
0J L 123/11; Guidelines on the application of Art. 81 of the EC Treaty to technology
transfer agreements, [2004] 0J C 101/2.
3 Council Regulation (EC) No. 139/2004 of 20 January 2004 on the control of concentra
tions between undertakings (the EC Merger Regulation), [2004] 0J L 24/1; Guidelines
on the assessment of non-horizontal mergers under the Council Regulation an the con
trol of concentrations between undertakings, [20081 0J C 265/6.
4 DG Competition discussion paper an the application of Art. 82 of the Treaty to ex
/
82
clusionary abuses, December 2005 (http://ec.europa.eu/competition/antitrust/art
discpaper200s.pdf), Guidance an the Comrnission‘s enforcement priorities in applying
Art. 82 EC Treaty to abusive exclusionary conduct by dominant undertakings, 3 December
2008 (http://ec.europa.eulcompetitionlantitrustlart82/guidaflCeen.Pdf).
5 Community framework for State aid for research and development and innovation,
13 December 2006. [2006] 01 C 323/1.
212
Kupzok, Sturny-Luder and Surblyte
IICVoL41
of unilateral practices by dominant firms. As a main concern in this respect
he identified the prevention of abuses of dominance without chilling compe
tition on the merits. Polo suggested solving this problem by using what he
calls an identification approach. First, a possible anticompetitive story is
identified which fits the main factual elements of the case and spells out its
empirical predictions. Secondly, an alternative competitive explanation and
its empirical predictions are developed. In a further step, the two alternative
stories are to be compared. In doing so, one can distinguish on the one hand
those factual elements which are consistent with both stories and on the
other hand those elements which are predicted only under one alternative
and thus allow to discriminate between the two stories. The application of
this approach (which would have led to a different outcome of a case) was
further illustrated by the presentation of the Italian RDB case
6 on selective
price cuts in the construction material industry.
Economic issues in antitrust were further explored by Daniel L. Rubinfeld,
the Robert L. Bridges Professor of Law and Professor of Economics at the
University of Caiifornia, Berkeley. Due to the growing use of empirical
methods in competition law, three controversies, namely the issues of market
definition, unilateral effects and merger simulation, were addressed. As to
the determination of the relevant market, Rubinfeld pointed out that though
necessary for unilateral conduct analysis (Art. 82 EC, Section 2 Sherman
Act), the market definition in some other cases (e.g. Art. 8 1(1) EC, Section 1
Sherman Act) is overly demanded by the courts and is frequentiy unneces
sary. Furthermore, increasingly sophisticated methods for the definition of
markets, such as cointegration methods, arc used by the economists. How
ever, such methods as the cham of substitution analysis can be criticised,
because almost anything can be qualified as a substitute, depending on how
the criteria are set. For unilateral effects Rubinfeld pointed out that,
although it is clear that market power is a prerequisite of the assessment, the
problem is that we still do not know how much market power is needed. lt is
also not clear how useful empirical methods could be. Moving on to merger
simulation as a tool to evaluate unilateral effects, a number of difficulties
involved were highlighted such as understanding demand, the making of
assumptions on price-cost issues and the use of simulation modelling. How
ever, Rubinfeld conciuded that although merger simulation requires strong
assumptions, it remains a valuable tool to evaluate unilateral effects.
Oliver Budzinski, Associate Professor in the Department of Environmental
and Business Economics at the University of Southern Denmark in Esbjerg,
focused on problems and limitations of using industrial economics in compe
tition law. First of all he pointed out that the application of sophisticated
(empirical) methods of industrial economics requires extensive and high
quality data. Secondly, competition agencies and courts may have an unin
tended focus on price-elements since modern industrial economics mainly
6 Italian Competition Authority, Provvedimento n. 17522, 24.10.2007.
2/20 10
Foundations and Limitations of an Economic Approach
213
focuses on (short-run) price and quantity effects, whereas non-quantifiable
and long-run effects may wrongly be ignored. Thirdly, competing economic
models brought by parties into proceedings may leave the courts with the
difficult task to identify the “right“ modei. However, this issue could be
mitigated by setting standards for economic expertise and by appointing
neutral non-partisan experts. Fourthly, the use of industrial economics in
competition law can lead to an unintended weakening of antitrust enforce
ment, as new theories and tools do not always fit into the “old“ institutional
environment. Furthermore, the predictive methods like merger simulation
arc based on a continuity assumption, whereas it is in fact unclear whether
the underlying competition process will remain unchanged post-merger. In
particular, mergers in narrow oligopolies as weil as abuses of dominant
positions can cause structural interruptions on the market. Finally, the use of
tools such as simulation models, econometrics and case-specific modelling
creates significant costs which must be compared to the benefits invoived. As
a result, a considerable part of the addressed limits of modern industrial
economics refers to an (overly-extensive) case-by-case approach. However,
Budzinski suggested that insights from modern industrial economics can be
used to shape better competition rules and presumptions.
Comment and Discussion
In the comments to the three presentations, Laurence Idot, Professor at Uni
versity Panthon-Assas (Paris II), agreed with the problem of data availability,
as weil as costs and time for coilecting it. She further expressed concerns that
not all areas of competition iaw receive the same level of attention, perhaps
because not all practices arc equally well suited for economic analysis. She
emphasised that it is not only important to develop economic tools, hut that it
is also crucial to control the reliability of such tools.
In the subsequent discussion, the question was raised as to how far economic
externalities, such as the global economic crisis, can affect and influence
industrial economics. lt was argued that the analyticai tools of economics
were capable of predicting most of the problems which arose in the global
economic crisis. Rather, the issue was that regulators were not receptive. The
argument that the change of environment was followed by a change of tools
was supplemented with the statement that over time not only the environ
ment, but rather our focus changed (e.g. on Art. 82 EC or merger cases),
which leads to the question why this shift in focus occurred. lt was further
pointed out that experience with a specific practice may lead to the conclu
sion that a simple rule previously applied proves to be wrong and that a
more complex standard for assessment is necessary. However, this raises
concerns as to the predictability. In order to improve predictability, the
development of a common understanding of which behaviour is more ac
cepted was considered to be more effective than safe harbours, since it may
not always be clear-cut whether a practice falls within the safe harbouz
Although the discussants agreed on the benefits of using economics in the
deveiopment of competition poiicy, it was admitted that there will always be
Kupzok, Sturny-Luder and Surblyte
214
IICVoI.41
a “gap“ between economics and the law: no legal rule, wide as it might be,
will ever fit the underlying economic theory 100%. The raised concern that
competition policy could become a victim of economic fashion (e.g., chan
ging attitude towards resale price maintenance) was rejected with the argu
ment that there is a greater risk that valuable new ideas never enter competi
tion law and practice than the danger of unfounded new ideas triggering
such changes. lt was also emphasised that economics plays an important role
not only in the handling of cases, but also in the setting of rules, as it can
provide structures on how to evaluate certain behaviour.
Panel 2: The Consumer Welfare Standard from an Economic
Perspective
Presentations
Professor Gregory Werden, Senior Economic Counsel in the Antitrust Divi
sion of the U.S. Department of Justice, Washington D.C., gave a presentation
on the meaning and role of “consumer welfare“ in competition law. Werden
first noted that the term “consumer welfare“ in today‘s discourse on competi
tion policy generally means consumer surplus. Re then pointed out that
“consumer welfare“ is generally used as the criterion for judging legality or
as a guide to the development of legal mies or to the application of the law in
particular cases. However, although consumer welfare could be regarded as
the ultimate goal of competition policy, Werden noted that by static analysis
of welfare we learn nothing about dynamic effects. Moreover, the ultimate
goal of merger control laws should be the promotion of social welfare, thus
the weffare of all consumers. For the question as to whether consumer
welfare should be applied as a guide or test for merger control, Werden
suggested that a total surplus standard could be a better test for assessing the
legality of mergers. Although merger control analysis should also include an
evaluation of long-term effects, the initial focus of such analysis should be on
short-term price effects, as they are still predominant. Similarly, the ultimate
goal of competition policy in the field of concerted practices and potentially
exclusionary conduct of dominant firms should be the promotion of the
welfare of all consumers. However, Werden pointed out that the effect on
“consumer welfare“ as such can neither be the appropriate test for legality,
nor an appropriate guiding principle for the application of the law. The focus
should rather be on the protection of the competitive process itself.
Viktor J. Vanberg, Director of the Walter Eucken Institute and Professor of
Economics at the Albert Ludwig University Freiburg, pointed out that a
“more economic approach“ raises two major issues: firstly, whether the goal
of competition policy should be the promotion of consumer welfare or total
welfare or rather the “freedom to compete“ (“Wettbewerbsfreiheit“); sec
ondly, whether an “effects-based“ or rather a “form-based“ approach should
apply. Re addressed these two issues from the perspective of constitutional
economics. Whereas the question which goal competition policy should
pursue is a normative question asking what ought to be (in the form of a
—
2/2010
Foundations and Limitations of an Economic Approach
215
categorical imperative or unconditional ought-statement), economics is an
empirical science which can thus provide advice only in the form of hypothe
tical imperatives (conditional ought-statements). In a democratic polity
policy advice to citizens should be supported by arguments that convince
citizens that following the advice supports their mutual advantage and their
common interest. This aspect has not been recognised by welfare economists,
who typically advise on aggregate social welfare effects, which entails that
sub-groups of citizens may suffer. The polity of a private law society commits
to respect the private autonomy of the individuals involved, including the
freedom to compete. Such freedom is restricted only through negative or
prohibition rules. With respect to the controversy on whether the goal of EU
competition policy should be “freedom to compete“ or “economic effi
ciency“, Vanberg pointed out that one should distinguish more carefully
between the constitutional level (choice of rules) and the sub-constitutional
level (choice within rules) of competition policy. Though efficiency or wel
fare considerations could be important at the constitutional level, applying
such considerations at the sub-constitutional level, i.e. to the judgement of
individual cases, would mean to subdue the exercise of economic freedom of
individuals to welfare considerations, which would be against the gmain of a
private law society. Finally, an “effects-based approach“, seeking to decide
each case on its own merits, may reduce legal certainty and would be
inconsistent with a private law system. The limits of our knowledge make
guidance through rules indispensible. A form-based or rule-oriented ap
proach allows us to compare pattemns of outcomes with rules. Economics
can give valuable knowledge on the constitutional level about how such rules
should be drafted.
Martin Hellwig, Director of the Max Planck Institute for Research on
Collective Goods and Professor of Economics at the University of Bonn,
elaborated on the normative foundations of competition policy by looking
more closely at the “cons“ of both welfare economics and freedom of
competition as goals of competition policy. With respect to the “cons“ of
welfare economics, he first of all noted that it is not clear what is meant by
the term “efficiency“. The mere focus of welfare economics on efficiency
considerations neglects the potentially severe impact of a practice or a
memger on distribution. Turning to the “cons“ of the freedom of competition
appmoach, Hellwig pointed out that the vagueness of the term “freedom to
compete“ is problematic. This approach ignores the need for a measuring
rod to appraise the relevant trade-offs between (i) inter-brand and intra
brand competition, and (ii) the rights of dominant companies, the rights of
competitors and the rights of suppliers or buyers. Although the preservation
of the market structure cannot be treated as an absolute priority since a
dominant company‘s right to compete on the merits must be respected, Ger
man and European competition lawyers seem to assume too readily that
market power is the product of illegitimate behaviour and that dominant
firms engaged in abusive behaviour. In particular in case of ambivalent
practices, trade-offs must be considered. Finally, Hellwig suggested that a
216
Kupzok, Sturny-Luder and Surblyte
IICVoI.41
“consumer surplus“ standard, according to which a given practice is analysed
by looking at the effects on the weil being of the other side of the market,
could be a useful standard in attempting to understand competitive issues,
although even then some difficulties remain with judging the trade-offs
between consumer surplus today versus tomorrow, or between consumer
surplus upstream versus downstream.
Comment and Discussion
Appraising the presentations, Daniel Zimmer, Professor at the University of
Bonn and member of the German Monopolies Commission, pointed out that
there is still no international consensus as to what the term “consumer
welfare“ exactiy means. According to Zimmer, “consumer benefit“ could
serve as a proxy for competitive forces. Re also cautioned that the focus in
merger control only on short-term effects, even if they are predominant, is a
risky approach. With regard to an efficiency or welfare economics approach
he agreed that it tends to be technocratic and dirigiste. Furthermore, he
raised the question as to whether there is an intra-economic explanation why
a consumer welfare (or benefit for the opposite side of the market) decision
rule should apply. With respect to the freedom of competition approach,
Zimmer pointed out that freedom of competition also encompassed freedom
of contract. Therefore, one should be cautious in restricting it.
The role of consumer welfare in competition policy was also the main topic
of the subsequent discussion. The need to differentiate between the ultimate
objective of competition policy and the applicable standard which the
authority shall apply in practice was highlighted. lt was first of all suggested
that a consumer welfare standard should be used in practice not because it
is the ultimate competition policy goal, but rather because it would be too
complicated to apply a total welfare approach. Furthermore, the treatment
of consumer welfare in the GlaxoSmithKline case as an objective of competi
tion law was considered to be incorrect. lt was further argued that the goal
of competition law cannot be consumer welfare since competition policy has
to respect and protect individual freedom. In connection with the competi
tion policy debate it was pointed out that the existing link between corisumer
welfare and freedom to compete on the one hand and fairness as weil as
distribution issues on the other hand should also be addressed. A further
topic discussed was the application of a rule-oriented versus case-by-case
approach. Since we can only take imperfect decisions, the application of a
case-by-case analysis was rejected. Rather, the use of rules which are not too
complicated to apply in practice was supported.
—
Panel 3: Normative and Institutional Limitations to a More
Economic Approach
Presentations
Professor Joachim Bornkamm, Chief Justice of the VI Civil Court of the
German Federal Supreme Court, opened the panel by expioring the norma
2/2010
Foundations and Limitations of an Economic Approach
217
tive and institutional limitations to a “more economic approach“. As it is
clear that legal rules have to meet a certain level of clarity, the application of
a “more economic approach“ is ambivalent. In fact, the notion of a “more
economic approach“ takes only some of the relevant issues into account, for
example, the application of the law, but not the circumstances when the
competition authority can step in. Moreover, even a higher standard of
clarity is needed if the norms are directly applicable, like Art. 81 EC and
Art. 82 EC. Such a need for legal certainty is particularly obvious in practice.
Although private enforcement is only a secondary way of enforcing competi
tion law norms, in Germany civil courts often have to answer the question of
validity of a contract under Art. 81 EC. Though rather rare under Art. 81
EC, damage or injunctive relief cases under Art. 82 EC are often dealt with
in civil courts. Thus, on the one hand, competition rules are based (and,
therefore, dependent) on economics which renders the economic reasoning
behind these rules relevant. On the other hand, even though the use of
economics can be useful in competition law, per se rules in some areas (e.g.
resale price maintenance) are important. Similarly, ciear-cut rules are neces
sary for horizontal agreements. Article 82 EC, however, is more open to
direct economic influence: while the obstruction in these cases is usually
easily identified, the analysis of possible justifications often proves to be
more difficult. This is especially true on the example of the ECJ judgments
concerning intellectual property rights.
Dirk Schroeder, partner at the law firm Cleary Gottlieb Steen Hamilton in
Cologne and Honorary Professor at the University of Cologne, analysed the
impact of a “more economic approach“ on the judicial control and certain
practical limitations to this approach. Competition law limits economic
activity, which is generally protected by different fundamental rights. This
occurs in, at least, three ways: by requiring prior authorisation of mergers,
by prohibiting the abuse of a dominant position and by prohibiting agree
ments that restrict competition and whose positive effects do not outweigh
negative effects. Though crucial in such cases, the role of the judicial control
risks to be reduced under the application of a “more economic approach“.
On the one hand, it could tip the balance in favour of the competition
agencies, which possess certain discretion with respect to assessments of an
economic nature. On the other hand, the European experience so far seems
to be encouraging, as the courts in fact exercise full, instead of a limited,
control. As to the practical limitations to a “more economic approach“, he
observed that in contrast to the past, when legal definitions were created by
the legislature or the courts, under a “more economic approach“ economic
assessment becomes part of the legal rule. This raises the danger of extended
agencies‘ power, as the judges may thus rely on guidelines issued by competi
tion authorities. In turn, this could enhance the role of economic experts, as
the judges might no longer feel qualified to apply the rules without expert
assistance. This could further lead to considerations as to the differing
qualifications of experts, the correctness and appropriateness of the tools
they apply, or even the “delegation“ to them to decide cases.
21
Kupzok, Sturny-Luder and Surblyte
IICVoI.41
Franois Souty, Senior Counsel to the Director General of Competition,
Consumer Affairs and Fraud Repression in the French Ministry of Economic
Affairs and an Associate Professor at the University of La Rochelle, ex
pounded on two types of limitations: normative limitations, due to divergent
views on economics, and institutional limitations, due to the different object
and subjects of competition policy (EU/U.S.). With regard to the divergent
views on the economic objectives and the economics of competition law
from the 1920s till 2009, four “schools of antitrust“ developed in the U.S.
over the 2Oth century were distinguished, namely: the Harvard School (fol
lowed by the “Structuralist“ and then the Post-Chicago School), the Chicago
School, the Industrialists (later “Evolutionists“) and, finally, the Nihilists or
Schumpeterians (or “Public Choice“). Normative limitations are evidenced
by the main points of economic divergence between the mainstream
“schools“, for example the definition and relevance of “barriers to entry“
analysis, different views of market definition, an Opposition between the
“workable competition“ and “contestable competition“ paradigms etc. In
stitutional differences exist mainly due to different object and subjects of
competition law in the EU and in the U.S. Within the EU, the object of
competition policy includes prohibitions of private anticompetitive beha
viour as weil as provisions on the control of State aid. Thus, here the subjects
of competition law are both private and public practices. In the U.S., “antitrust“ concerns exclusively private sector behaviour and violations of antitrust law are treated as criminal and civil law offences, as opposed to
European competition law which treats them mainly as administrative law
offences.
Comment and Discussion
The comment on the three presentations was delivered by Oliver Budzinski.
Addressing the aspect of predictability, he agreed on the advantages of
predictable rules (inter alia, reduced transaction costs), but pointed out that
any change of die law reduces predictability. In fact, a phase of reduced
predictability is acceptable if it is necessary to improve rules. Due to the
growing role of economic expertise, he emphasized that there are many other
areas where judges rely on experts. Thus, it is hard to agree with the
argument that the application of a “more economic approach“ would reduce
judicial control and enhance the power of agencies. Though a “more eco
nomic approach“ indeed increases the scope of appreciation, the application
of guidelines should not enhance the role of agencies, as guidelines, after all,
remain non-binding.
In the following discussion, the scope of appreciation of judges was first of
all discussed. lt was admitted that, although the appiication of a “more
economic approach“ can reduce the scope of appreciation of judges, an
opposite experience could be observed in Europe. Furthermore, the danger
that judges will “delegate“ cases to experts is not necessarily a consequence
of a “more economic approach“ and thus not an argument for or against it.
The need though could arise to train the judges. With regard to a “more
2/2010
Foundations and Limitations of an Economic Approach
219
economic approach“ being a part of the legal rules it was noted that in
competition law, from the very beginning, there is less certainty, first and
foremost, because of its “functional approach“. Thus exactly this has to be
reflected and interpreted with the help of legal norms. lt was further cau
tioned that the new EC Guidelines on non-horizontal mergers, which adopt
a “more economic approach“, could pose bigger problems in five to ten years
when at that point other concepts of which we may not be aware of today
may be interpreted into the guidelines. Attention was also drawn to the
normative foundations of competition law, for example die fundamental
rules anchored in the German Constitution on establishing a market econ
omy. Finally, the need to bear in mmd that economic theory changes and
that there is still no convincing tool to develop competition policy further
was emphasized.
Panel 4: Fresh Blood for Competition Law
Evolutionary Economics
—
Innovation and
Presentations
Howard Shelanski, Professor of Law at UC Berkeley, focused on merger
policy and innovation. With an approach he developed together with Profes
sor Michael Katz, which attempts to fit the concept of innovation into the
basic framework of merger policy, Shelanski exposed two common prob
lems, namely the role of market definition and the treatment of efficiencies.
Innovation can be central to merger policy in two ways: tbrough the Innova
tion Impact Effect, and the Innovation Incentives Effect. In the former,
innovation can so strongly affect the relationship between the pre-merger
and post-merger marketplace that the current circumstances could only serve
as a poor predictor of future conditions. In the latter, innovation can be of
itseif an important criterion to assess economic progress. Shelanski further
pointed out that although there is theory and evidence that support the view
that short-run prices tend to rise with concentration, the evidence on the
relationship between concentration, innovation and welfare is much less
cicar and interpretations more complex. Two linkages, namely between
concentration and innovation and between welfare and innovation need to
be assessed. The former is wealdy supported by ambiguous theory and facts:
while on the one hand competition itseif can spur R&D investment, it is on
the other hand argued that large firm size and high market shares are
conducive to R&D investment, due to the fact that big sunk costs necessitate
high returns to allow cost recovery. Empirical studies do not resolve this
ambiguity. The welfare-innovation linkage is also ambiguous. While in
theory, firms may invest more than the socially efficient amounts in R&D, in
practice, private incentives for R&D typicaliy are too bw. Furthermore,
small changes in innovation can have very large welfare effects, so that
efficiencies are difficult to be projected. As a result, the need to change
current standards of merger assessment and burden of proof could arise. In
this respect, it was proposed to adopt a flexible approach to market defini
220
Kupzok, Sturny-Luder and Surb1yt
IICVoI.41
tion, to take seriously the complexity of the concentration-innovation link
age and to treat uncertainty using standard toois of decision theory.
Uwe Cantnei Professor of Microeconomics at the Friedrich Schiller Univer
sity of Jena, dealt with industrial dynamics and innovation within evolution
ary economics. A certain composition of an industry or market and the
related performance of firms therein is dependent on the path along which
these structures emerged. Thus, a line of research focused on the knowledge
and capabilities of actors and firms, as weil as on the search and learning
processes involved in building them up. The focus on the underiying capabil
ities for innovation requires a consideration of their behaviourai foundations
and their embedment in the technological environinent. Using the OACK
(opportunities, appropriability, cumulativeness, knowledge) approach, Cant
ner illustrated how ways in which agents bring about new ideas vary.
Namely, the degree of differences depends on the four OACK features, for
example on the availability of the technological opportunities that the com
pany is able to harvest and economize on. In this aspect, uncertainty of
success and risk of investment piay a significant role and imply that a simple
caiculation of an expected vaiue of the returns is not applicable. Thus, the
concept of dynamic capabilities, which introduces long-term strategic plan
ning into the agent‘s behaviour, couid be inciuded in the analysis. Further
more, the combination of the elements of the OACK approach with the
behaviourai foundations for innovative activities wouid aliow for the hetero
geneity of firms in a sector to be reiated to differences in knowledge and
competences built up over time. In this sense, the knowledge and compe
tences of firms are a major source for industrial structures and their change
over time.
Wolfgang Kerber, Professor at the Philipps University of Marburg, ex
pounded on competition, innovation and maintaining diversity through
competition and focused on competition as a process of parallel experimen
tation and the importance of multiplicity and diversity of parallel research.
Kerber argued that though more assessment is necessary on the effects of
mergers, agreements and other business practices on innovation (Shelanski‘s
view) and to extend the theoretical basis of competition policy beyond main
stream, game-theoretic industrial economics (Cantner‘s view), it should be
noted that many theoretical and empirical insights of these approaches have
not been used so far in competition law practice so that much more research
in regard to both innovation and how it shall be applied in competition law
is needed. Thus, Kerber elaborated on the specific economics of parallel
processes, or to put it differently, on the merits of diversity which means that
a more diverse Pool of resources, such as products or technologies, might
increase a firm‘s probability of being capable of responding quickly to
unanticipated exogenous shocks, as weh as enhance the endogenous capabil
ity to develop better solutions to existing problems. Also, availability of
multiple independent sources of innovation might be important for rendering
the market effective as a selection device. One of the advantages of diversity
and parallel experimentation is that a Pool of different technologies could be
2/2010
Foundations and Limitations of an Economic Approach
221
regarded as a Pool of options. Furthermore, he investigated whether diversity
and parallel experimentation should be protected through competition law.
On the one hand, more parallel innovations (more trials) have overall
positive effects, on the other hand, the quality of trials can suffer if too many
are conducted. Thus the core question would be what the optimal number of
parallel researches (“optimal diversity“) is, which cannot be precisely an
swered at present. Kerber concluded that, although merger and R&D agree
ments could lead to negative effects on competition (e.g., reduction of
parallel innovations), factors on which they depend have not yet been
accounted for in the current research.
Comment and Discussion
Appraising the presentations, Andreas Heinemann, Professor at the Univer
sity of Zurich, with regard to general remarks on evolutionary economics
observed that the focus should be on the competitive process: after all, ideas
are tried in the market, so that it cannot be predicted at the outset whether
such ideas, or which of them, will survive. In this way, the objective of
competition policy is to maintain the competitive market structure. Further
more, Heinemann raised the question as to whether the “4plus“ test should
constitute and be further promoted as a safe harbour. Finally, he addressed
the concern as to how innovation could be better integrated into Art. 82 EC,
in particular, bearing in mmd the Microsoft case, where the court had
rejected the argument of Microsoft that the making available of interface
information to competitors would reduce its own innovation efforts. The
question thus arises as to whether the argument of innovation could be used
in favour and on the part of a dominant firm or rather on a whole industry.
The following discussion focused on three issues, namely considerations of
market definition, standardisation and apphication of evolutionary econom
ics. Regarding market definition, the difficulty of defining the relevant
market in the hight of innovation was chahlenged, basically due to the
constantly changing nature of the relevant products which could further lead
to a too broad or a too narrow market definition. However, it was acknowl
edged that market definition is not comphicated in all cases concerning
innovation. The standardisation issue was brought up to suggest that stan
dards lead to development of just one technology, thus contradicting the
assumptions of the option theory. In this vein, it may be so that even though
standards do have certain advantages (e.g., the reduction of transaction
costs), they also have the potential of preventing future progress of innova
tion, in particular in the longrun. This raises the question as to whether
competition for the market or rather competition within the market should
be preferred in such cases. Also, the need to maintain diversity was high
highted. Though admitted that in markets with network externahities the
setting of standards might reduce diversity, it was argued that the possibility
of diversity could be maintained in the processes before setting the standard.
Finally, it was noted that in the European Commission‘s approach to the
innovation incentives of Microsoft and its competitors, potential can be seen
222
Kupzok, Sturny-Luder and Surblyte
IICVoI.41
for the use of evolutionary economics, as the European Commission uses
implicit evolutionary economic arguments in its decision. lt was argued that
focusing on short-term effects may be a mistake, but in order to incorporate
long-term effects more research would be needed.
Panel 5: Fresh Blood for Competition Law Behavioural,
Experimental, and New Institutional Economics
—
Presentations
Christoph Engel, Director of the Max Planck Institute for Research on
Collective Goods, Bonn, and Professor at the University of Bonn, expanded
on behavioural economic explanations of competition and cartelisation by
presenting a meta-study, which he had conducted on oligopolies and degrees
of collusion. The evidence used for the study included 147 experimental
papers, which were published until 2006 on oligopoly experiments with
510 measurable parameter constellations. This evidence was rendered com
parable by generating two indices: the deviation from the Walrasian Equilib
rium and the Nash Equilibrium. From the formal model, three benchmarks
were calculated: the Competitive Walrasian, the collusive Pareto, and the
strategic Nash Predictions. As a result, Engel could propose that the degree
of collusion between economic actors can have either a strategic or a beha
vioural expianation, and he suggested that the theory of conditional coop
eration could be a promising explanation for varying degrees of collusion.
According to the presented concept, persons can be classified as conditional
cooperators if they are willing to contribute to the public good as long as the
other group members contribute as weil. By mapping the oligopoiy problem
to a linear public good, Engel revealed certain differences between a stan
dard linear public good and an oligopoly and pointed out the need of further
experiments in order to show whether the existing differences matter and
whether the concept of conditional cooperation can be used for oligopolies.
Professor Bart Wilson from the Economic Science Institute at Chapman
University, Orange, California, anaiysed how experimental economics can
be used as a tooi to understand market competition. He defined experimen
tal economics as a laboratory method of inquiry for studying how indivi
duals interact in controlled settings defined by a specific set of rules. Wilson
pointed out that experimental economics can help shed light on a wide range
of antitrust issues. This was evidenced by the presentation of Wilson‘s recent
research on the controversial topic of zone pricing in gasoline markets,
which is a practice whereby refiners set different wholesaie prices for retail
gasoline stations depending on the geographic area. In an experiment with
two geographical areas, an isolated and a clustered one, both a zone pricing
situation and a uniform pricing situation were simulated, applying computer
software which could perfectly enforce both pricing strategies. Wilson de
monstrated that uniform wholesale prices hurt consumers since, according
to the resuits of the experiment, consumers in clustered areas paid higher
prices when zone pricing was banned, as compared to when it was allowed.
2/2010
Foundations and Limitations of an Economic Approach
223
In contrast, consumers in isolated areas paid the same price irrespective of
whether zone pricing was allowed or banned. According to Wilson, the
reasons for this result are twofoid. First, high refiner prices for isolated areas
are not the cause for higher station prices in isoiated areas, but rather the
consequence thereof. Station prices in isoiated areas are higher than in
ciustered areas because consumers prefer not to travel long distances in order
to buy at iower prices and because there is only one local gasoline station.
Second, uniform wholesale pricing results in the unintended consequence of
refiners setting the uniform price levei generaliy higher than the comparable
zone price in the clustered area. Drawing conclusions from the Zone pricing
experiment, Wilson suggested that well-meaning interventions of competi
tion authorities, which are designed to manipulate market allocations, can
backfire, because the authorities cannot account for the complex incentives
in an intricate industry.
Justus Haucap, Professor at the Friedrich Alexander University of ErlangenNürnberg and Chairman of the German Monopolies Commission, dealt
with bounded rationality and competition poiicy. Re defined bounded
rationality broadiy, as any deviation from perfect rational behaviour, and
proposed that it should serve as a link between Behavioural Economics and
New Institutional Economics. The central question in Haucap‘s presentation
was thus how competition law shouid be designed in order to take into
account the bounded rationality of both competition authorities and courts
in deciding competition cases, on the one hand, and of consumers on the
other hand. With regard to the former, Haucap highlighted the procedural
aspects in designing competition law and emphasized the aspect of evidence,
especially the question on how much information the authority needs to
collect in a proceeding and who carries the burden of proof. Re then pointed
out that competition rules should be designed so as to minimize the sum of
expected welfare costs from type-I (false positives) and type-Il errors (false
negatives), as weil as information and other transaction costs. As to the
bounded rationality of consumers, he argued that behavioural economics
can yield interesting insights, especialiy because final consumers are usually
not as rational as assumed by neociassical theory. As an illustration, Haucap
presented the evidence-based examination carried out during the German
Monopoiies Commission‘s action against Vodafone and T-Mobile on com
plaint from the entrant E-Plus. Thereby, an experiment was carried out to
model consumer behaviour concerning the complaint, which was based on
the market foreciosure argument in which E-Plus argued that the combina
tion of tariff mediated network effects and first-mover advantages (large
instalied base) had been used to leverage market power. A survey carried out
among 1,000 students in order to analyse the relationship between mobile
teiecommunications prices and customer behaviour revealed that consumers
appeared to make substantial mistakes in choosing tariffs. Thus, beha
vioural economics may lead exactly to the opposite evaluation of certain
pricing strategies than a more traditionai view with perfectly rational con
sumers.
224
Kupzok, Sturny-Luder and Surblyte
IICVoL41
Comment and Discussion
Delivering the comment, Professor Andreas Fuchs from the University of
Osnabrück recognised that modern economic concepts introduce more fea
tures of real life into competition law and thus help better understand how
markets work. They may serve as a blueprint for redesigning the competition
law concept. Nevertheless, he drew attention to the fact that experimental
economics is limited to certain assumptions, as it is conducted in the labs,
which neglect a multitude of real market conditions. While the crucial point
in the experiment is the design of markets and economic environments, in
fact, only parts of the environment can be taken into account and other parts
remain unknown. Thus, experimental economics cannot be regarded as an
exclusive instrument for assessing economic behaviour.
In the subsequent discussion, first the concept of collusive behaviour was
scrutinized and it was recognised that a certain problem remains in that
while in experiments individuals interact, such interaction in the market
takes place between firms. Therefore, it is not possible to elaborate antitrust
rules based only on experiments. Yet, it was commented that many relevant
questions only arise once experiments are actually conducted. lt was con
cluded that it is not possible for competition authorities to rely on single
experimental evidence. Collection of evidence is crucial for the decision of a
competition case. With regard to the concept of collusion, it was pointed out
that oligopoly practices may fall under both Art. 81 EG and Art. 82 EG and
thus it is necessary to consider whether “collusion“ as examined in experi
ments covers abuses of dominance under Art. 82 and not only cartels and
concerted practices under Art. 81 EG. lt was also cautioned that taking into
account bounded rationality of consumers could be a too far-reaching ap
proach.
Panel 6: The Political Economy of an Economic Approach to
Competition Law
With the focus on how improvements in institutional design and institutional
capacity can increase the ability of a competition agency to operate effec
tively, Professor William E. Kovacic, Commissioner of the US Federal Trade
Commission, listed at least three mainstream questions that competition
authorities face: firstly, why and how much a jurisdiction should rely on
competition (instead of alternative models) to organise economic activity;
secondly, which economic ideas should underpin the design and application
of competition laws; and, thirdly, which approaches to public administration
increase the legitimacy of choices among alternative economic ideas. As
competition is not the only possible model of economic organisation, the
basic question of political economy is why competition should be promoted.
Bearing in mmd that, in the past, the competition model was suppressed
time and again, typically in the face of crisis, Kovacic deemed it important to
offer persuasive reasons why reliance on market mechanisms and competi
tion law is the correct approach. Whereas it is undoubted that economics do
2/20 10
Foundations and Limitations of an Economic Approach
225
matter in competition law, the question remains to what extent. This leads to
the assessment as to which economic ideas should shape competition policy.
In any case, the choice of economic techniques cannot be decided without
careful assessment of how they will be applied. Furthermore, as competition
agencies enj oy discretion and a margin of appreciation in applying economic
theories to specific cases, it becomes important to increase the legitimacy as
to the choices made. The question as to how best to engineer competition
policy was surveyed in the FTG‘s Self-Study (“The FTC at 100“), which
aimed to identify areas in which the FTC could improve the institutional
foundations on which it operates and tried to identify ways as to how to
overcome the dilemma in politics of short-term appointments and conse
quential underinvestment in long-term capabilities. The general outcome of
the Self-Study was a redefinition of what a “good“ competition agency is
and what characteristics of good administrative practice it should possess.
Josef Drexl, Director at the Max Planck Institute for Intellectual Property,
Gompetition and Tax Law, Munich, and Honorary Professor at the Ludwig
Maximilian University of Munich, focused on the political dimension of the
economic approach to competition law, placing it in a larger context than
economics and the law, and referred to Adam Smith the pioneer of modern
economics, but, above all, a moral philosopher. The analysis of the political
character of the economic approach was based on the appraisal of the
objectives of competition law, and was then followed by the assessment of
theoretical shortcomings of the economic approach and finally explored in
the light of normative values. Drexl argued that, although economic effi
ciency, as the ultimate goal of competition policy, is treated as promoting
welfare of society at large, this view seems to be very collectivist. Modern
societies, by contrast, are built on the political will of individuals rendering
the political support for the law dependent on the acceptance of the law by
each individual. Furthermore, since individual competition law decisions
take from some and give to others, concerns as to the “distributive justice“
arise. Similarly, the second objective of competition law, namely the promo
tion of consumer welfare, at first glance presupposes making everybody
better off. Yet, several political arguments against “consumer welfare“ can
be raised. First of all, since “consumer welfare“ is only used as a criterion for
measuring efficiency, the same criticism as for the efficiency standard applies.
Secondly, the application of a consumer welfare standard would require
predictions on the effects of a certain conduct. However, on the basis of
Hayek‘s definition of competition as “a discovery procedure“, it is clear that
often only the competitive process itself reveals consumer interests, which
are thus rather unpredictable. For example, the European Commission in
Microsoft was only able to demonstrate consumer preference for certain
products, because the market had already produced that information.
Thirdly, the problem of conflicting consumer interests arises, in particular
where R&D and IPRs are involved which may promote the interests of some
consumers but restrict those of others. The economic approach can, further
more, suffer from several shortcomings, for example, the economic approach
—
226
Kupzok, Sturny-Luder and Surb1yt
IIC Vol. 41
could be regarded as being not “economic“ enough or even leading to
inconsistent policies. Finally, the political dimension of the economic ap
proach could be explored in the light of some normative values, above all,
equal treatment and economic freedom. The principle of equal treatment
would mean equal opportunity for any additional competitor to enter (how
ever, not to stay in) the market, thus maintaining markets open and increas
ing consumers‘ ability to make their own choices. By contrast, allowing firms
to rely 011 an efficiency defence and thus to escape competition law would be
a violation of this fundamental principle. By the same token, the efficiency
defence can lead to a mistaken concept of freedom to compete.
Discussion
In the subsequent discussion it was added with regard to merger control that
the protection of the competitive process in the long run is much more
important and sustainable than a possible lowering of prices as a result of
the merger in the short-run. Furthermore, efficiency as the goal of competi
tion law was criticised in the sense that if competition policy used the
standard of efficiency, which possibly conflicts with the principle of equal
treatment, this might cause resistance by the electorate and would hence be
inappropriate. After all, competition is an institution that serves everybody
(like a common good). Moreover, the application of very simple economic
models could open the door to bring in political convictions. The need was
raised for better political explanations of a “more economic approach“
which would fit the current societies as well as reflect the goals of the EG
Treaty and in particular the Lisbon Treaty which, in effect, advocates for
human rights and equal treatment. Thus, though general economic theory
could, indeed, help better understand how markets work, its application
within competition law should not be treated as an absolute concept. lt was
furthermore suggested to shift to a different kind of consurner protection
within competition law: instead of a mere focus on lowering of prices, it was
proposed to focus on consumer sovereignty meaning that consumers have
more choice. Finally, the question was raised as to how economics should be
used in competition law: whether it should be applied only in particular
fields such as mergers, or whether it should be involved in the designing of
appropriate rules such as a structured rule of reason. The need was also
emphasied to recall that according to the doctrine of separation of powers
the addressee of the question as to what aims competition law should pursue
is the legislature rather than the European Commission or the courts.
Ciosing the conference, Josef Drexl referred to the quotation of Confucius
used as his opening remark that “real knowledge is to know the extent of
one‘s ignorance“ and expressed his hope in that the conference showed the
lack of knowledge as to the particular aspects of a “more economic ap
proach“ which would result in a more cautious application of it.